Harvard Law Development

By Tom Watts*

This is the third in a series of blog posts previewing some of the conceptual questions that may appear in the same-sex marriage oral arguments on Tuesday. Part I is here, and Part II is here.

In previous posts, I discussed whether same-sex marriage bans classify by sexual orientation. If they do, the next question the Supreme Court must answer is what level of scrutiny a sexual orientation classification should receive. Race classifications receive strict scrutiny, which almost always means the law in question is unconstitutional. Sex classifications receive intermediate scrutiny, which often means the law is unconstitutional. Most other classifications receive only rational basis review, which almost never means the law is unconstitutional, unless the Supreme Court determines that the law was enacted due to “animus” (here, anti-gay bias).

As a technical matter, the Supreme Court has laid out various standards — to pass strict scrutiny, a law must be narrowly tailored to a compelling government interest and must be the least restrictive means of achieving that interest, whereas to pass rational basis review, the law need only be rationally related to a legitimate government interest — but because the Supreme Court has provided only cursory explanations of the general principles of this analysis, it is hard to explain precisely what this means. How do we know whether something is a legitimate (or compelling) government interest? Are there legitimate government interests that are not compelling government interests? The Court’s decisions do not answer these questions clearly. Indeed, one of the most prominent lower-court judges, Judge Posner of the Seventh Circuit, showed some impatience with the sheer conceptual confusion around these standards, and the Supreme Court has the option of trying to clarify, although no one on the Court has indicated a willingness to do so.

It is likely that the marriage bans will be struck down under any legal standard, but if the Supreme Court decides that sexual orientation classifications deserve intermediate scrutiny, that decision could reverberate in LGBT rights cases throughout the country. Any anti-gay law in any state would suddenly be vulnerable to serious challenge. On the other hand, if sexual orientation classifications receive only rational basis review, the Supreme Court is likely to find that the marriage bans were enacted with animus. Consequently, plaintiffs challenging anti-gay laws would have to find evidence of some kind of anti-gay bias in the enactment of the laws in order to get them struck down. It is likely that more laws would survive under this framework than under intermediate scrutiny.

How will the Supreme Court decide the level of scrutiny? There are five factors that sometimes arise, which are phrased in different ways (and not all are always invoked). One factor is a history of discrimination against the group. A second is that the group is obvious and distinct (in legalese, “discrete and insular”): group membership can be determined at sight (discrete) and group members are in contact with each other significantly more than with non-group members (insular). A third is the political powerlessness of the group, whether because it is a demographic minority (e.g., African Americans are, and have always been, much less than half of the population of the United States) or because it is underrepresented in the halls of power (e.g., women). A fourth is often referred to as “immutability,” but because classifications by religion receive strict scrutiny and religion is mutable, it might be more accurate to describe this factor as “fundamental to identity.” A fifth is that the group’s defining traits are generally irrelevant to the group’s ability to participate in and contribute to society.

The opposing briefs argue whether gays and lesbians meet these criteria (for what it’s worth, it seems clear to me that they do), but the broader question is what the test itself consists of. A series of decisions from the late 1960’s to the mid-1970’s recognized new protected classifications that trigger heightened scrutiny (race, sex, and several others), but the Supreme Court appears to have lost its appetite for protecting new groups. It has been decades since the last classification was recognized (arguably, illegitimacy in 1977 or 1978). If the Supreme Court seriously considers protecting sexual orientation, the way different Justices phrase and discuss the factors may be telling. How immutable does the trait have to be? How do we know when a group is politically powerless? There are many questions here that the Supreme Court could answer.

Also, even if the Supreme Court doesn’t recognize sexual orientation discrimination as a type of sex discrimination, as discussed in a previous post, holding that sexual orientation classifications receive intermediate scrutiny would have nearly the same effect, because that is the level of scrutiny that sex classifications receive. The plaintiffs and the federal government have both argued for this position. It is rare for the government to support recognizing a new protected classification, so the government’s position may carry particular weight with the Court. Then again, the federal government made the same argument in Windsor, and the Court refused to bite, so the Court may duck this issue again.

If the Supreme Court does in fact recognize a new protected classification for the first time in decades (and, perhaps, clarify the legal test for protected classifications), its doing so might be the most legally significant aspect of the decision. This is worth watching closely in oral argument.
* Tom Watts is a fourth-year JD/MPP student at HLS and HKS and is the Executive Online Editor of HLPR.

By Tom Watts*

This is the second in a series of blog posts previewing some of the conceptual questions that may appear in the same-sex marriage oral arguments on Tuesday. Part I is here.

In the previous post, I discussed the prospect of the Supreme Court deciding that a state same-sex marriage ban was a sexual orientation classification on the face of the law. In this post, I’ll consider the route to concluding that the bans should receive scrutiny as sexual orientation classifications even though they do not explicitly classify by sexual orientation. As I’ll explain, if this topic comes up, it may be one way of determining where the Justices stand on the questions presented in the cases.

Ordinarily, the question that courts ask about a law that doesn’t classify in its text, but that has some kind of disparate impact on one group over another, is whether the law was enacted with a discriminatory purpose. Here, the fact that gays and lesbians are barred from marrying a partner of their choice (but straight people aren’t) isn’t enough unless the Court finds that these laws were enacted with gay people in mind. Of course, most of the specific provisions at issue actually were; after Hawaii appeared on the verge of recognizing same-sex marriage in the 1990’s, there was a wave of statutory enactments (including DOMA) to bar same-sex marriage in most states, and after Massachusetts did recognize same-sex marriage, there was another wave of state constitutional enactments. The legislative history of these enactments is clear enough about the anti-gay intentions of the laws.

But there’s a wrinkle here. The Supreme Court didn’t explicitly take up the question of whether Ohio, Michigan, Kentucky, and Tennessee’s bans were unconstitutional. It took the question, “Does the Fourteenth Amendment require a state to license a marriage between two people of the same sex?” For the answer to be “Yes,” it seems as though the Court would have to decide that all existing (and perhaps any conceivable) legal barriers to same-sex marriage are unconstitutional. Even if the court strikes down the statutory and constitutional provision barring same-sex marriage, common-law rules that existed before those provisions still barred recognition of same-sex marriage. Did those laws develop with discriminatory purposes? Common-law principles barring same-sex marriage date at least as far back as medieval England, prior to the invention of the concept of sexual orientation in the nineteenth century (perhaps due to an increasing medicalization of the understanding of sexuality). The rules can’t have developed with discriminatory purpose if no one conceived of the group as existing at the time.

There’s a possible way out: the very first time that the Supreme Court found a violation of the Equal Protection Clause in a law that did not include a facial race classification, it did so when the disparate impact was truly egregious. A San Francisco city ordinance forbade the operation of a laundry in a wooden building without a permit. Every permit application from a Chinese person was denied, and nearly every permit application from a white person was granted. There was no need to find intent in the enactment of the law because the administration of the law was so clearly discriminatory; hence, the law was struck down. Here, too, there may be no need to find express homophobia in the common-law development of principles that forbade same-sex marriage, because a same-sex marriage ban has an equally extreme disparate impact: every marriage license application from a gay person is denied, and almost every application from a straight person is granted.

The case is not perfectly analogous: the wooden-laundry law gave officials discretion which they exercised in a discriminatory manner, whereas there is no comparable discretion in the marriage bans. If anything, though, this makes the marriage bans even more unconstitutional. At least San Francisco could have implemented the law in a way that was not discriminatory; there is no such option for the marriage bans.

Is the Supreme Court likely to raise this point? Probably not in the “licenses” question, quoted above, for two reasons. The first reason is the criminal law. Enacting a law that turns out later to have a large disparate impact can’t always be unconstitutional, or else nearly all of the criminal law would be unconstitutional, because nearly all of it has a large disparate impact on African Americans (consider Blewett). There is a way out: the Supreme Court could hold that an unconstitutional disparate impact is one that is absolute (a license or permit is either granted or not) and nearly uniform in its disparity between the two groups (basically every member of one group gets the license or permit, and basically every member the other doesn’t). This would save the criminal law. However, to my knowledge, no one has made this argument.

The second reason this point is unlikely to arise in the “licenses” question is the opinion below. Judge Sutton, who wrote the opinion in the court from which this case is being appealed, raised the point of the common-law nonrecognition of same-sex marriage. However, he did so to argue that the constitutional bans were not motivated by animus, which is clearly wrong. The legislative history of these enactments, beginning in 2004, shows plenty of explicit anti-gay sentiment.

If the Court does start raising questions about the common-law history of marriage, it may be because the Court is interested in the second question it granted: “Does the Fourteenth Amendment require a state to recognize a marriage between two people of the same sex when their marriage was lawfully licensed and performed out-of-state?” Many states, such as New York, recognize marriages performed out-of-state without determining whether those marriages could have been performed in-state. As a result, a ban on recognizing out-of-state same-sex marriages would be contrary to New York’s usual approach. (New York enacted same-sex marriage legislatively, so this is no longer an issue in New York specifically, but in any other state with a similar approach, this point remains true.)

In contrast, according to Ohio’s brief (pp. 32-33), Ohio recognizes out-of-state marriages that would have been valid or voidable under Ohio law, but not marriages that are void under Ohio law. As a result, not recognizing same-sex marriages in Ohio is consistent with Ohio’s usual approach: such marriages are void under Ohio law anyway, and Ohio doesn’t recognize any out-of-state marriages that are void under Ohio law.

The upshot of all this is that if the Court were to answer (in the first question) that a state does not have to grant marriage licenses, it could answer (in the second question) that those states that have very strong comity principles, as New York does, are required to recognize out-of-state same-sex marriages, but those that take Ohio’s approach are not required to do so.

It will be worth listening closely to the Supreme Court’s oral argument to determine whether the Justices seem interested in any of these issues, because it may give an indication where they are on the two questions.

For more detail on several of the points in this post, see Zachary Robert Herz, The Marrying Kind. The author thanks Zach for his help in developing much of the content in this post. Of course, any errors should be attributed to the author alone.

* Tom Watts is a fourth-year JD/MPP student at HLS and HKS and is the Executive Online Editor of HLPR.